[Steven Stahlberg]

Escher experimented with some weird distorted perspective grids, don't remember exactly what though.

[Dan]

Hey, I'm just wondering if there are any writings on the subject of curved perspective? ie a perspective system in which the horizon line curves all the way around 360 degrees. I recently did some studying and wrote up a little presentation on it, but I'm wondering if the stuff I did is actually original or if some other people have already done this in the past.

[Ian Jones]

I think its called curvilinear perspective.. and you'll find it at least mentioned in most perspective books. I would imagine it not to be tricky, but just incredibly tedious.

Actually I would be interested to read your presentation, I want to know more about perspective. Can you put it online?

Escher is defintely 'the man' when it comes to stuff like this.

[Dan]

well it wasn't all in digital form. But I'm sure if you know how regular perspective works you can understand this. Just imagine that your horizon line wraps completely around. So it's 360 degrees. Because it is 360 degrees, this means you can't represent it on a flat piece of paper. It will only be truly correct if you wrap the paper into a circle around your view point, that's your eye. A straight line (running parallel to the ground) represented on your paper will begin at a point somewhere on the horizonline and end at the point opposite it(that's half the horizon away). So that means that this line has to curve to get between the points unless the line happens to lie exactly on the horizon line. The actual curve of the lines to fit between two points is a sin wave. It's kind of hars to explain why it is without some examples and demonstrations, but it is a sin wave.

[Dr. T.J. Eckleberg]

Best way to illustrate this is to render a 360� panorama in a 3D program. Here is a quick one out of Bryce 4.



Because the human eye acts more like a movie camera than a still camera, curvilinear perspective is closer to the way we perceive convergence of parallel lines. Linear perspective looks right, however, because the eye is accustomed to converting "on the fly" as we rest our gaze on a succession of points.

[Dan]

I'm not sure how to classify the human eye. In the end it does work like a camera with a little hole to let light through, but is the back part where the light is recieved curved? But in the end it all doesn't matter because your mind interprets all the information into a 3 dimensional image in some very complex ways.

[Dan]

anyways, back to the original post, has there been any major writings or past stuff on curvilinear perspective